
Interview Framework
Before Claude writes a spec or delegates to a subagent, this skill runs a structured interview to gather requirements through dialogue. It follows a three-phase algorithm: understand the problem by asking clarifying questions (always leading with a recommended answer), propose 2-3 concrete approaches with honest trade-offs, then confirm and store decisions in .progress.md. The codebase-first exploration rule is smart: it uses the Explore agent to discover facts from code before asking you anything, so you only answer questions about preferences and priorities. It also watches for completion signals like "done" or "proceed" to skip ahead when you're ready. Works across all Ralph phase commands that need adaptive conversation before execution.
- GitHub stars
- 340
- last updated
- May 19, 2026
Install
$ npx -y skills add tzachbon/smart-ralph --skill interview-framework --agent claude-codeInstalls into .claude/skills of the current project.
Files
Interview Framework
Treat every normal-mode interview governed by this framework as a grill. Run the approval-gated interview for start, triage, research, requirements, design, and tasks. Treat this skill and its references as the single source of truth for interview behavior. Phase commands supply exploration territory and artifact context; they do not redefine the algorithm.
Quick mode bypasses interview questions only. It still requires current discovery, contract loading, bypass receipts, delegation checks, and artifact-agent load parity.
Entry Contract
Before each new or resumed interview:
- Complete the applicable skill discovery pass from
${CLAUDE_PLUGIN_ROOT}/references/normal-mode-gates.md. - Reload this entire
SKILL.md,references/algorithm.md,references/domain-modeling.md, every selected skill body, and every selected skill resource required for the current work. Loadreferences/examples.mdonly when an example is needed. - Record the load manifest with
phase_gate.py record-skill-load. - Begin or resume the interview with the matching phase, interview ID, discovery revision, and context digest.
Sponsors
Details
- Repository
- github.com/tzachbon/smart-ralph
- Category
- AI & Agent Building
- License
- MIT
- First seen
- Jun 11, 2026
More from tzachbon/smart-ralph
All 28- MCP Integration
- Plugin Settings
- Plugin Structure
- Ralph Specum
- Ralph Specum Cancel
- Ralph Specum Design
- Ralph Specum Feedback
- Ralph Specum Help
- Ralph Specum Implement
- Ralph Specum Index
- Ralph Specum Refactor
- Ralph Specum Requirements
- Ralph Specum Research
- Ralph Specum Start
- Ralph Specum Status
- Ralph Specum Switch
- Ralph Specum Tasks
- Ralph Specum Triage
- Reality Verification
- Skill Development
- Smart Ralph
- Spec Workflow
- Speckit Workflow
- Command Development